Many entities require data replication. Such replication becomes more difficult as the databases associated with such entities increase in size.
The most basic method of data replication is mirroring data from a central data processing station to a safe site. The safe site preferably preserves data when the service interruptions occur at the central data processing station. Such service interruptions may include power outages associated with natural disasters, failures at the central data processing station or other service interruptions.
The mirroring of data from the central data processing station to the safe site often takes time. When a service interruption occurs at the central data processing station, the lag time between data being received at the central data processing system and the mirroring of the data to the safe site complicates a determination of as to which data was successfully mirrored to the safe site prior to the service interruption. Preferably, all data must be reconstructed to ensure that no data was lost as a result of the service interruption.
It would be desirable to provide an end-to-end architecture and method that allows cataloging of, and recovery of, preferably all data in the system at the time of a service interruption.